The exponential growth in terms of wireless data traffic along with the number of mobile devices is expected to continue. For example, according to Cisco, it is predicted that the number of mobile devices will reach 1.5 per capita in 2021; there will be 11.6 billion mobile-connected devices including machine-to-machine modules. Global mobile data traffic will increase sevenfold between 2016 and 2021. On the other hand, the energy consumption of cellular networks world-wide per year is around 60 billion kWh and is expected to double within the next few years. Specifically, the large portion (80%) of electricity used in cellular networks is consumed by base stations, emitting over a hundred million tons of carbon dioxide annually. Combining the need for high wireless data rates, the increase in power prices and the raising environmental concerns, cellular network providers are facing unique challenges leading to huge increases in capital and operating expenditures. As a result, the need for wireless networks meeting the ever increasing demand in wireless data in a sustainable way is more pressing than ever. To achieve sustainable energy and spectrally efficient wireless networks, several promising technologies, such as massive multiple-input multiple-output (MIMO), millimetre wave communication, network densification and energy harvesting, are under investigation. Cloud radio access network (CRAN) is considered as a potential solution to achieve network densification, and hence will meet the exponential growth in wireless network traffic, in a cost and energy efficient manner. CRANs facilitate increase in network capacity and energy efficiency while reducing both network capital expenditure and operating expense. However, we strongly believe that the potential of CRANs is still not fully exploited and their performance can significantly be improved via incorporating new technologies such as massive MIMO and energy harvesting. The aim of this project is to design sustainable high energy and spectral efficiency CRANs. The novelty of this project is that we propose and optimise a new CRAN architecture incorporating a massive MIMO central unit and remote radio heads (RRHs) equipped with hybrid energy sources (i.e., with energy harvesting capability and connection to the power grid) to further improve the energy and spectral efficiencies of wireless networks while being sustainable. To achieve the goal of this project, we will consider both long-term and short-term performance optimisation of the proposed CRAN architecture. In terms of long-term performance, given the flexibility in placing the RRHs offered by using wireless fronthaul, we propose to optimise the placement of the RRHs. In terms of short-term performance, we propose to design a low complexity channel estimation method and to jointly design the fronthaul and access links. The optimised design of the new CRAN architecture will lead to sustainable and improved energy and spectral efficiency wireless networks.

The Internet of Things (IoT) has digitally transformed our everyday life with exciting applications such as smart home, connected healthcare, smart cities, manufacturing automation, relying on billions of devices that have become connected over the past decade. Ofcom estimated that the number of IoT devices in the UK will soar from 13 million in 2016 to 156 million in 2024. Low power wide area networks (LPWANs) are new IoT systems with features of low power and wide coverage (over several kilometres). LPWAN accounts one-fourth of the number of IoT devices and the market. Digital Catapult is building the national LPWAN to improve the qualities of our lives and boost the UK economy using LoRaWAN, NB-IoT, and SigFox technologies. Vodafone and Three UK are piloting NB-IoT for a nationwide cellular-based LPWAN. However, this digital revolution can only be viable if we can provide secure wireless connections. A pair of keys should be established between legitimate devices for encryption and decryption prior to transmissions. Although conventional key distribution schemes, e.g., elliptic curve Diffie Hellman (ECDH) algorithm, are quite mature, they tend to be less suitable for lightweight IoT applications owing to their high complexity. In practice, the pre-shared key (PSK) is often used, which may never refresh the key after its initial configuration. This obviously presents security risks since the key can be revealed, e.g. by side channel attacks. What is worse, many users lack awareness. The UK's first cyber survey in 2019 by the National Cyber Security Centre revealed numerous weak passwords, e.g., 23.2 million victims worldwide used 123456 as their passwords. The vulnerabilities of IoT have resulted in numerous grave security attacks, which have compromised user privacy, adversely affected the economy and undermined the trust in the society. Gartner reported the information security market exceeded $124 billion in 2019. Indeed, conceiving secure yet low-complexity key distribution for low-cost IoT devices is challenging. It becomes even more difficult and cumbersome if key refreshing is needed, such as in LPWAN where IoT devices are located in a far-flung environment over several kilometres radius and may not be attended. This open challenge can be tackled with a radical and completely different approach, namely key generation from wireless channels, which automatically generates cryptographic keys from unpredictable characteristics of the wireless channel, and thus avoids the conventional key distribution. While key generation has been demonstrated to work well with short-range communications such as WiFi, its exploration with LPWAN technologies such as LoRa and NB-IoT is rather limited, due to the more complicated radio propagation conditions and the affected channel reciprocity. This project hence will bridge this gap by designing scalable, automatic, and lightweight key generation solutions for LPWAN. This project will be the first systematic study for LPWAN-based key generation. A synergistic approach will be adopted which involves theoretical modelling, algorithm design and experimental validation. The core aspects of this project will include novel mathematical channel correlation models and channel decomposition algorithms as well as new key generation protocols tuned and optimised for LPWAN. Extensive field-measurements will be carried out to evaluate our algorithms. A unique feature of this project will be the creation of viable security solutions for IoT validated by extensive measurements and proof-of-concept prototypes.

The idea is to design a "dialogue system" interface to existing databases of the arguments surrounding controversial topics such as "Should the United Kingdom remain a member of the European Union?" or "Should all humans be vegan?". In particular, a user can have a "Moral Maze" style chat with the dialogue system. "Moral Maze" is a longrunning popular BBC 4 Radio programme in which a panel discusses a controversial topic with the help of witnesses and a host who chairs the conversation. The dialogue system consists of a panel of Argumentation Bots (ArguBots) who present arguments for or against the topic under discussion (the pro and con ArguBots), a host ArguBot and a witness ArguBot (that can provide detailed evidence). The user is invited to join the panel and voice their views on the topic under discussion. Thus the user can explore what they thought and what others thought about the controversial topic. An important part of the projects will be to evaluate the effects on people's appreciation of the complexity of debate and attendant ability to comprehend the world from other people's point of view or perspective.

Spin injection and transport in semiconductors is under intense investigation by physicists around the world, motivated by fascinating new insights into condensed matter, aware of considerable potential for novel devices and ensuing technologies. However, spin injection and its detection pose exceptional challenges. Much focus has been on technologically important materials: GaAs, where optical properties aid spin detection, and more recently Si for its long spin lifetimes. Here, we propose a new approach based on germanium. Ge is compatible with Si technology, has a longer spin life time than GaAs, a higher room temperature hole mobility than GaAs or Si, and better modulation properties than Si due to its higher spin-orbit coupling. SiGe heterostructure technology also has the potential to increase spin diffusion lengths by virtue of dramatic enhancements in carrier mobility. We recently carried out optical experiments that demonstrated RT spin transport and extraction through Ge for the first time, based on a structure consisting of Ge grown epitaxially on GaAs and an electrodeposited Ni/Ge Schottky contact [C. Shen et al., Appl. Phys. Lett. 97, 162104 (2010)]. Here, we propose to build upon that work and use the Si-Ge system to its full extent, through delta doping and bandstructure-engineering to maximize spin transparency of the electrical contacts and using strain and low dimensionality to enhance coherent transport in the channel. The culmination of this project should be the exciting prospect of the elusive two-terminal semiconductor spin valve operating at room temperature and an early demonstration of spin modulation by a gate electrode in such a device. The programme will combine the complementary expertise of the partners: Warwick in SiGe epitaxy and in carrier transport, Southampton in Schottky barrier research, and Cambridge in semiconductor spin transport by optical and electrical means, together with the facilities of the Southampton Nanofabrication Centre and industrial support from Toshiba Europe Research Ltd.

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.